The main goal of this study was to examine the effect of different viewing environments on the performance of a spatial perspective-taking task. The spatial perspective-taking task is an egocentric task in which viewers are asked to change their perspective mentally and point to a target from an imagined perspective. The task requires egocentric encoding (the stimuli are encoded in relation to the viewer’s body) and transformation (the viewer imagines own body changing its orientation to face a new perspective). Twenty-four participants were administered a computerized perspective-taking task presented in three different viewing environments of increased immersivity: traditional computer desktop (2D), three-dimensional large projector screen (3DPS), and fully immersive virtual environment (3DVE) offered through the use of the head-mounted display.
Participants exhibited a similar linear dependence of accuracy and response time on the magnitude of imagined heading in all three conditions. Furthermore, all three conditions were not significantly different in their response time or accuracy. However, as immersivity of an environment increased, the number of egocentric errors, related to the confusion between left and right, front and back, also increased. In contrast, the number of adjacent errors, representing allocentric strategies and related to under-rotating or over-rotating the objects, decreased. The results suggest that although large projector screens provide a higher degree of immersivity than conventional 2D displays, in contrast to fully immersive 3DVEs, they are insufficient to ensure the dominant use of an egocentric, viewer-centered frame of reference.
Keywords
AUGMENTED AND VIRTUAL REALITY (AR/VR)
Additional Keywords
perspective taking, immersivity, egocentric, allocentric, spatial frames of reference, virtual reality